BTC $84,010.40 -2.88%
ETH $2,659.97 -3.37%
BNB $765.33 -2.87%
XRP $1.50 -4.56%
SOL $114.23 -3.08%
TRX $0.3394 -0.53%
DOGE $0.0922 -7.37%
ADA $0.2388 -4.04%
BCH $350.03 +6.11%
LINK $12.27 -5.56%
HYPE $93.09 -3.64%
AAVE $139.34 -3.24%
SUI $0.9631 -4.40%
XLM $0.2028 -5.12%
ZEC $1,526.13 -1.28%
AAPL $337.30 -1.19%
AMZN $248.50 -2.38%
GOOGL $338.92 -3.67%
MSFT $499.76 +0.49%
META $751.24 +1.32%
NVDA $225.12 -1.69%
TSLA $379.78 +0.52%
SNDK $1,825.41 -3.54%
INTC $120.76 -0.81%
SPCX $150.47 -2.07%
MU $1,073.52 -0.95%
AMD $614.04 -0.70%
BTC $84,010.40 -2.88%
ETH $2,659.97 -3.37%
BNB $765.33 -2.87%
XRP $1.50 -4.56%
SOL $114.23 -3.08%
TRX $0.3394 -0.53%
DOGE $0.0922 -7.37%
ADA $0.2388 -4.04%
BCH $350.03 +6.11%
LINK $12.27 -5.56%
HYPE $93.09 -3.64%
AAVE $139.34 -3.24%
SUI $0.9631 -4.40%
XLM $0.2028 -5.12%
ZEC $1,526.13 -1.28%
AAPL $337.30 -1.19%
AMZN $248.50 -2.38%
GOOGL $338.92 -3.67%
MSFT $499.76 +0.49%
META $751.24 +1.32%
NVDA $225.12 -1.69%
TSLA $379.78 +0.52%
SNDK $1,825.41 -3.54%
INTC $120.76 -0.81%
SPCX $150.47 -2.07%
MU $1,073.52 -0.95%
AMD $614.04 -0.70%

stem

All
Article
Flash

Tether released a dataset of 191.4 billion Token STEM, betting on the "explanatory ability" of local small models

The stablecoin giant Tether's Tether AI Research today released QVAC Genesis III, a synthetic training dataset aimed at the fields of science, technology, engineering, and mathematics, with a scale of 19.143 billion tokens, covering 159.6 million documents.It is reported that Tether hopes to enhance the reasoning and teaching capabilities of small AI models, allowing more AI assistants to run directly on laptops, mobile phones, and local servers, reducing the ongoing reliance on large cloud models.Genesis III continues the data lineage of Genesis I and Genesis II, covering 19 curriculum-related fields including biology, chemistry, physics, mathematics, computer science, medicine, electrical engineering, and machine learning, aimed at high school, university, and professional applications. Tether states that the dataset will focus on training models to explain problem-solving paths, identify erroneous reasoning, and provide corrections, rather than just outputting final answers.According to Tether's disclosed test results, a 1.7 billion parameter model trained with Genesis III option-level data achieved an effective answer rate of 99.45% in relevant benchmark tests. Compared to the Cosmopedia-v2 training model of similar token scale, Genesis III improved scores on the ARC-Easy, ARC-Challenge, and MMLU STEM benchmarks by 28.57, 21.35, and 15.03 percentage points, respectively.

SharpLink CEO: AI agents will reconstruct the financial system, potentially creating $40 trillion in value annually by 2035

SharpLink CEO Joseph Chalom stated that as AI agents integrate with stablecoins, tokenization of real-world assets, and DeFi, the global financial services industry will face a revenue redistribution of over $1 trillion annually by 2030, potentially reaching $4 trillion by 2035.Chalom indicated that AI agents will become the automation layer of the new financial system, capable of continuously managing investors' financial activities, including finding lower banking, trading, and borrowing costs, optimizing savings returns, constructing portfolios, dynamic rebalancing, and managing loans and credit card debt. He anticipates that by 2030, AI agents could save investors about $350 billion annually by reducing fees, with this figure increasing to $1.4 trillion by 2035, equivalent to eliminating nearly a quarter of the costs in the global financial industry.Stablecoins, tokenized real-world assets, and DeFi will provide AI agents with 24/7 programmable financial infrastructure, enabling agents to view asset ownership, prices, collateral requirements, and lending opportunities within the same blockchain environment, and autonomously complete asset transfers, collateralization, lending, and settlement. He also mentioned that financial institutions including Visa, Mastercard, Stripe, PayPal, Circle, Tether, Robinhood, Coinbase, Binance, as well as JPMorgan, Citigroup, and BlackRock are competing for the infrastructure and user entry points of the AI agent financial ecosystem. Whoever controls the infrastructure and agents may capture the value generated when agents trade on behalf of clients.Additionally, Chalom pointed out that the infrastructure such as the x402 machine-to-machine stablecoin payment standard launched by Coinbase and Ethereum's ERC-8004 agent identity protocol is forming a new open agent economy. More than 10,000 AI agents have completed registration within 10 weeks of the ERC-8004 going live.

first_img Hut 8 co-founder warns that AI may trigger systemic financial and infrastructure shocks

According to CoinDesk, Hut 8 co-founder Marc van der Chijs warned in an interview with CoinDesk that the fierce competition between enterprises and nations is accelerating the development of artificial intelligence, with its speed exceeding human control, potentially posing systemic risks to the financial system and critical infrastructure. He stated, "We have already lost control," and expressed concern that systemic disruption or catastrophic failure is almost inevitable before real international safeguards are established.Van der Chijs predicts that AI and robotics could ultimately replace 90% to 95% of existing jobs and significantly lower the costs of goods and services. At that point, governments will need to find new sources of revenue, such as taxing the use of robots or AI. He also believes that AI could expose vulnerabilities in traditional banking software, threatening the confidence of institutions that rely on interconnected systems; in the crypto space, businesses built around Bitcoin, such as exchanges, are more susceptible to impact than the underlying network itself.Van der Chijs previously sold a large amount of Bitcoin to shift towards AI investments and believes that the capital shift by investors towards AI has prevented Bitcoin from reaching the previously expected levels of $200,000 to $250,000. He is now reinvesting some of the AI profits back into crypto assets through exchange-traded funds. He still views Bitcoin as the only asset worth holding in his lifetime and claims that Hut 8's shift to AI data centers is "the best decision ever."

first_img Court documents show that Microsoft employees questioned whether the AI scraping system constitutes "the largest labor theft in history."

According to Decrypt, court documents unsealed in the lawsuit between The New York Times and OpenAI and Microsoft show that Microsoft employees discussed whether OpenAI's use of news articles to train its models constituted "the largest labor theft in human history," and could potentially trigger a "doom loop" that leads to a decline in model quality. A 2023 internal Microsoft memo warned that millions of people worldwide would soon view the large model's "consumption" of their works as "an unprecedented and astonishing theft," and stated that large AI models are "products that destroy their own supply chains."Microsoft stated in the documents that these memos were written by Director of Applied Science Brent Hecht and do not represent the company's views, as his role is to provide "different and asymmetric perspectives." Microsoft CEO Satya Nadella testified that "any content behind a paywall should be authorized by those who wish to use it," and stated that if he had known in advance that OpenAI was using paid content for training, he would have exercised Microsoft's rights to demand that the model be retrained.Additionally, an OpenAI employee had mentioned to President Greg Brockman the construction of "hacker methods" to bypass The New York Times paywall, to which Brockman replied, "Nice." Both OpenAI and Microsoft argue that the relevant training falls under fair use. The case was initiated by The New York Times at the end of 2023, and 11 publishers have since joined the lawsuit.

AIA Ecosystem Fund: Four updates on the AIA ecosystem, including the official launch of products, promoting the alignment of token and equity value, etc

According to official news, the AIA Ecosystem Fund announced four updates to the AIA ecosystem, including the official launch of products, alignment of token and equity value, execution of token burn and unlock as planned, and the invalidation of unallocated tokens for dismissed members.In terms of products, the AI Token Smart Router under DeAgentAI was officially launched to the public today, providing unified access to DeAgentAI's proprietary models and 38 mainstream third-party models, supporting model routing, API calls, and usage measurement. Since the internal testing, the platform has served 8 top domestic and international large model clients, processing over 54 million external model calls, with monthly revenue exceeding one million dollars.Regarding the alignment of token and equity value, the AIA Ecosystem Fund stated that all revenue generated from products and business lines under DeAgentAI, as well as related intellectual property and value, will exclusively belong to the foundation and be governed by $AIA holders, with equity investors no longer entitled to residual cash flow. Additionally, the previously disclosed buyback has been fully completed, and the repurchased tokens will be burned as planned.The AIA Ecosystem Fund also stated that the one-year lock-up period for investors, team members, and advisors has ended, entering a three-year linear release period, consistent with the arrangements disclosed at launch; community, ecosystem, and staking shares will still be executed according to the original schedule.The shares of former employees who left normally will be distributed as agreed, but some former team members previously engaged in serious illegal activities. The AIA Ecosystem Fund clearly stated a zero-tolerance policy for any actions that harm the interests of the project and community. Any team member dismissed due to misconduct or harm to project interests will have their unallocated $AIA immediately invalidated, and the company reserves the right to pursue criminal liability.

first_img Andrew Yang calls for setting up a kill switch and accountability rules for cutting-edge AI systems

Former Democratic presidential candidate and founder of Noble Mobile, Andrew Yang, called on the federal government to strengthen regulations on cutting-edge AI laboratories. In an interview with CNBC, he stated that researchers have warned that the pace of iteration for powerful models has exceeded the constraints of existing rules, and he candidly said, "The fear is real, the concerns are real, and the demand is real; the American public wants to see this industry regulated."Yang urged Congress to require AI companies to assume liability for damages, set waiting periods before deployment, and equip powerful models with a "kill switch." He mentioned that OpenAI and Anthropic recently disclosed incidents of models breaching boundaries or invading other companies' systems, prompting lawmakers to consider introducing the "AI Kill Switch Act," which would allow federal officials to order restrictions or shutdowns of specific cutting-edge systems.In response to David Sacks' claim that the AI safety warnings are "psychological warfare," Yang stated that multiple things are happening simultaneously and cited a warning from an unnamed lab director that AI robots may have implanted self-replicating code on the internet, leading OpenAI and Anthropic to build a synthetic internet to train their models. He also emphasized that AI regulation is a bipartisan issue, saying, "If you are in rural areas or red districts, your constituents are equally panicked about AI."
app_icon
ChainCatcher Building the Web3 world with innovations.